Safety and Efficacy of Axatilimab at 3 Different Doses in Patients with Chronic Graft-Versus-Host Disease (AGAVE-201)

Background: Chronic graft-versus-host disease (cGVHD) is an immune-mediated complication of allogeneic hematopoietic cell transplant (alloHCT) that affects multiple organs with inflammatory and fibrotic pathology, leading to significant patient burden and mortality. Colony-stimulating factor 1 receptor (CSF-1R)-dependent monocytes and macrophages play a key role in cGVHD inflammation and fibrosis. Axatilimab (SNDX-6352) is an investigational, high-affinity anti-CSF-1R monoclonal antibody that targets monocytes and macrophages. We previously demonstrated the biological and clinical activity of axatilimab with organ-specific responses and symptom improvement in a phase 1/2 study (NCT03604692) in patients with cGVHD. Methods: This pivotal phase 2, open-label, randomized, multicenter study evaluated axatilimab at 3 different doses in alloHCT patients with recurrent or refractory cGVHD (AGAVE-201; NCT04710576). Eligible patients were randomized 1:1:1 to receive intravenous (IV) axatilimab at 0.3 mg/kg every 2 weeks (Q2W), 1 mg/kg Q2W, or 3 mg/kg every 4 weeks (Q4W). Randomization was stratified by severity of cGVHD and prior use of ibrutinib, ruxolitinib, or belumosudil. Concomitant use of corticosteroids, calcineurin inhibitors, or mammalian target of rapamycin inhibitors (sirolimus or everolimus) was allowed. Axatilimab treatment could be continued as long as there was clinical benefit as assessed by the investigator. The primary efficacy endpoint was overall response rate (ORR) in the first 6 cycles (24 weeks) as defined by the NIH 2014 consensus criteria; the efficacy boundary of the ORR is based on the lower bound of the 95% CI exceeding 30%. The key secondary endpoint was the proportion of patients reporting a clinically significant reduction of symptoms, as measured by the modified Lee Symptom Scale (mLSS) score with a threshold of ≥7 points. Safety endpoints included frequency and severity of treatment-related adverse events (TRAEs) and treatment-emergent adverse events (TEAEs). Results: A total of 241 patients were enrolled across 121 study sites and 239 (99.2%) patients were treated with axatilimab at the data cutoff of 7 April 2023. Patients were heavily pretreated with a median of 4 prior lines of therapy, including ruxolitinib (74%), belumosudil (23%), and ibrutinib (31%). Demographics and disease characteristics were balanced among the 3 dose cohorts (Table 1). ORR (95% CI) was 74% (63, 83) with 0.3 mg/kg Q2W, 67% (55, 77) with 1 mg/kg Q2W, and 50% (39, 61) with 3 mg/kg Q4W (Table 1). Median duration of response (DOR) has not been reached in any of the cohorts, with 60%, 60% and 53% of patients maintaining response at 12 months in the 0.3 mg/kg Q2W, 1 mg/kg Q2W, and 3 mg/kg Q4W dose cohorts, respectively. Clinical benefit, as measured by reduction in mLSS score, was reported in 55%, 54%, and 36% of patients in the 0.3 mg/kg Q2W, 1 mg/kg Q2W, and 3 mg/kg Q4W cohorts, respectively, in the first 6 cycles. Drug discontinuation owing to TEAEs occurred in 6% of patients with 0.3 mg/kg Q2W, 22% with 1 mg/kg Q2W, and 18% with 3 mg/kg Q4W. The most common TEAEs are summarized in Table 2. Fatal TEAEs occurred in 1.3%, 8.6%, and 7.6% of patients in the 0.3 mg/kg Q2W, 1 mg/kg Q2W, and 3 mg/kg Q4W dose cohorts, respectively. The frequency of TRAEs and grade ≥ 3 TRAEs were dose dependent, consistent with CSF-1R inhibition-mediated macrophage clearance. Most infections were mild, with 3 reported cytomegalovirus infections, including reactivations, in the higher dose cohorts. Conclusions: The AGAVE-201 pivotal trial met its primary endpoint for all doses studied. Axatilimabtreatment of refractory cGVHD in heavily pretreated patients resulted in robust clinical activity and durable responses in all 3 dose cohorts, with the highest ORR and least toxicity at the 0.3 mg/kg Q2W dose. Adverse events consisted primarily of transient laboratory abnormalities and other on-target effects known to be associated with CSF-1R block.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.